Implementing Deep Learning Models: a quick overview
During this talk I will provide a quick introduction to Deep Learning, starting from its theoretical foundations to simple implementations. Then, I will focus on some languages and frameworks for developing commercial projects, with a particular emphasis on Tensorflow and Keras. In particular, I will explain how these frameworks work by simple examples and fragments of codes. To conclude, some tips for feeding deep networks, and evaluate its performance are presented. Therefore, beyond to present a general overview on Deep Learning and its mathematical structure, this talk aims to indicate the first steps for build up a (simple) model in Python.
Marco Alberto Javarone is a Senior Researcher at nChain LTD and Associate Researcher at the University of Kent. He holds two PhDs, one in Computer Engineering and one in Applied Mathematics. His research activity is focused on Statistical Physics and Information Theory with applications to Machine Learning (and Deep Learning), Blockchain Technologies, Evolutionary Game Theory and Complex Networks. Recently, he published a book on computational models for Evolutionary Game Theory (Springer). Passionate for Startups, he works as consultant in the field of Machine Learning and Data Science.